Intrahepatic cholangiocarcinoma (ICC) is a highly malignant primary neoplasm that is considered to be clinically important and therapeutically challenging due to its increasing incidence, high mortality rates, and limited treatment options for patients who most often present with advanced fatal disease. A hallmark feature of ICC is a prominent desmoplastic stroma enriched in ?-smooth muscle actin-positive cancer-associated fibroblastic cells (?-SMA+CAFs). However, while it is becoming increasingly apparent that ?-SMA+CAFs may be playing a crucial role in promoting ICC progression, specific mechanisms through which such myofibroblastic-like cells may act to promote aggressive ICC remain elusive. A major development of the previous grant cycle was our establishment of an orthotopic syngeneic rat model of ICC progression that recapitulates key pathological, molecular, and clinical features of early and advanced stages of human desmoplastic ICC. Recently, we also succeeded in developing and partially characterizing a novel three- dimensional organotypic co-culture model of cholangiocarcinoma that complements our in vivo ICC model. Using co-culturing, we further showed ?-SMA+CAFs derived from orthotopic rat ICC to significantly promote cholangiocarcinoma cell ductal growth and invasiveness in vitro, as well as to stimulate up-regulation of specific genes associated with ICC invasive growth and progression (e.g., CXCR4, HGF, Muc1). We further demonstrated periostin, a matricellular protein synthesized and secreted by ?-SMA+CAFs in ICC, to be more highly expressed in rapidly growing, invasive ICCs than in slow growing, low invasive ICCs formed in our rat in vivo model. As a logical extension of these findings, we are now proposing two Specific Aims. Specific Aim 1 is focused on clarifying the functional role of periostin as a potentially important mediator of ICC progression. Under this aim, we will also test our hypothesis that ?-SMA+CAFs generate convergent proinvasive signals to cholangiocarcinoma cells through periostin/integrin ?4, HGF/Met, and SDF-1/CXCR4-mediated activation of FAK/PI3K-Akt/Rac1. Specific Aim 2 will utilize our unique rat organotypic cholangiocarcinoma cell culture and orthotopic ICC models, together with clinically relevant targeted agents, to preclinically validate an innovative moleculr strategy for ICC therapy based on combinational targeting of interactive ?-SMA+CAF/ cholangiocarcinoma cell pathways (e.g., hedgehog signaling pathway, HGF/Met, and SDF-1/CXCR4) associated with ICC progression. We anticipate the results generated by the proposed research to clarify the role of periostin in ICC progression and elucidate its relationship to select growth factor/chemokine-mediated signaling pathways by which ?-SMA+CAFs cross-talk with cholangiocarcinoma cells to facilitate aggressive malignant behavior. Moreover, we believe data generated from the proposed research will be of real value in identifying more potentially effective and rational treatment strategies for patients with progressive ICC, which hopefully will lead in a meanigful way to the development of new clinical trials.